Radar

N. N. Malov

The use of radio-engineering methods for detecting enemy aircraft, accurately determining their location, increasing the accuracy of antiaircraft-artillery fire, achieving exceptional precision in aimed bombing, correcting the firing of the navy, and a number of other important operations played an enormous role in the successful development of major bombing raids, in the defense of one’s own territory against enemy aircraft, and in ensuring the relative safety of sea communications.

The fulfillment of all these tasks required the creation of a new branch of radio engineering—radiolocation; the apparatus used for this purpose have recently received the name radars (the initial letters of the words “Radio Detection and Ranging,” i.e., “detection and determination of distance by means of radio”). The scale of production of radiolocation equipment may be judged from data cited in the journal Electronics, according to which, by July 1, 1945, the United States alone had “spent 2.7 billion dollars on the production of radars and half of that amount on their use.”

For understandable reasons, during the war there was almost no information on these questions for broad sections of the population. But in a number of recent issues of foreign scientific-technical and popular journals1–5 there have appeared articles devoted to radars (see the partial bibliography at the end of the article), which have also been used in the present survey, written in March 1946 and, naturally, in no way claiming to provide a complete account of radar technology, rightly called by some authors “the miracle of our time.”

1. Basic Principles of Radiolocation

The basic idea of radiolocation is extremely simple: if an electromagnetic wave sent into the atmosphere meets an aircraft in its path, it will be partially scattered by it; the energy reflected by the aircraft can be received by a corresponding receiver. Radio-

technical method of detecting an aircraft may have (and in fact does have) a number of advantages in comparison with former methods—searchlights and sound locators. Searchlights have a limited range of action, which is greatly reduced in foggy weather; they are completely ineffective in cloudy weather; a pilot who sees a searchlight can try to avoid its beam and, in any case, knows that he has been detected when he enters the region of action of the beam. Sound locators likewise have a limited range of action; their operation is complicated by the presence of extraneous noises, unavoidable under combat conditions, and by the difficulty of distinguishing the noises of enemy aircraft from those of one’s own. In mass raids and at the present speeds of aircraft they prove almost useless.

Fig. 1.

In addition to simply detecting an aircraft \(S\) in the air, a radar installation must make it possible to determine its coordinates accurately, i.e. the azimuth \(\alpha\) (Fig. 1), measured from some fixed direction \(OA\) (the observer is at the point \(O\)); the so-called angle of elevation \(\varepsilon\), which determines the altitude of flight; and the slant range \(R\) (in the case of locating naval vessels the problem is simplified, since \(\varepsilon = 0\)).

These tasks are solved by using directional antennas and by radiating into space not a continuous wave, but very short pulses, whose duration (as well as the pauses separating them) is determined above all by the range for which the radar is designed. Since the energy reflected from the aircraft and reaching the receiver constitutes an exceedingly small fraction of the transmitted energy (thus, for an operating range of \(150\) km it does not exceed \(10^{-17}\) of the radiated energy), at the moment when the reflected pulse is recorded the transmitter radiation must be absent. On the other hand, in order to avoid errors, the reflected pulse must return to the receiver before the next pulse is radiated. Thus the duration of the pause between pulses for a radar with a range of action \(R\) km must satisfy the condition

\[ \tau \geq \frac{2R}{c}, \tag{1} \]

where \(c\) is the speed of light.

As for the pulse itself, it must end before the signal reflected from an aircraft located at ...

at a minimum distance \(R_{\min}\), so that the pulse duration must satisfy the condition:

\[ \tau_1 < \frac{2R_{\min}}{c}. \tag{2} \]

In fact, the time interval during which reception of the reflected signal is impossible exceeds the duration of the emitted pulse because of the presence in the receiver of transient processes which continue, after the termination of the pulse, for a certain interval of time depending on the quality factor of the receiver circuits. Obviously, \(\tau\) and \(\tau_1\) constitute (taking into account the enormous magnitude of the speed of light) negligible fractions of a second. Thus, in the American SCR-268 radars, which were in service before 1944, the pulse duration is 5 microseconds, the pause duration is 235 microseconds, with a radar range of about 40 km and \(R_{\min} \approx 2.5\) km.

Since the fraction of the energy reflected by an airplane is vanishingly small, the pulse itself must have very considerable power, and the receiving device must have enormous sensitivity. The need to have directional radiation, however, compels one to strive for as high oscillation frequencies as possible.

Some idea of the choice of the parameters characterizing radar can be given by the following considerations. Let a transmitter, radiating uniformly in all directions, emit pulses of power \(P\) watts. On a target at a distance of \(R\) meters from the transmitter there will fall an energy flux of density

\[ S = \frac{P}{4\pi R^2}\ \text{watts}/\text{m}^2. \]

Since practical transmitters have directional action, then, assuming that the target is situated in the direction of maximum radiation, we obtain for the actual energy flux

\[ S_1 = SG_0, \]

where \(G_0\) is the so-called gain factor, which determines how many times the energy passing through an element of area located in the direction of maximum radiation exceeds the flux that would be obtained from a uniformly radiating transmitter with the same total radiated power. This factor in radar antennas ranges from 100 to 1200. Thus, for example, a directional transmitter with \(G_0 = 500\) and a power of 100 kW is equivalent to a uniformly radiating transmitter with a power of 50,000 kW.

A target reflecting a wave is equivalent to a new transmitter whose power is \(\sigma S_1\) watts, where \(\sigma\) is a multiplier depending on

sizes and shape of the reflecting surface. Of course, the target scatters energy differently in different directions, but for simplicity we shall understand by \(\sigma\) an averaged value for a uniformly scattering reflector. The scattered wave, arriving at a receiver located near the transmitter, has an energy-flux density

\[ S_2=\frac{\sigma S_1}{4\pi R^2}=G_0\sigma\,\frac{P}{16\pi^2 R^4}\ \text{watts}/m^2. \]

If the receiving antenna has an equivalent area \(A_2 m^2\), then it will absorb the power

\[ P_2=A_2S_2\ \text{watts}. \]

For antennas made in the form of reflectors, having as the aperture a circle of diameter \(D\) meters and feeding a dipole installed in the focal plane, we have

\[ A_2=\frac{\pi D^2}{4}\ m^2. \]

It is known that for such an antenna the gain is \(G_2=\frac{4\pi A_2}{\lambda^2}\), where \(\lambda\) is the wavelength used in the operation of the antenna. If we assume that transmission and reception are carried out on one and the same antenna, then, setting the minimum permissible received power \(P_2=P_{\min}\), we obtain from the preceding equations the following expression for the maximum operating range of the radar:

\[ R_{\max}=\sqrt[4]{\frac{PA^2\sigma}{P_{\min}4\pi\lambda^2}}. \]

This expression indicates that, for given antenna dimensions, it is advantageous to reduce the wavelength.

If, however, we introduce into consideration the width of the radiated beam \(b\), i.e., the angle between the directions in which, respectively, the maximum energy and an energy equal to \(50\%\) of the maximum are radiated, then, as calculation shows (for sufficiently narrow beams),

\[ b \simeq \frac{\lambda}{D}, \]

and for the operating radius we obtain

\[ R_{\max}=\sqrt[4]{\frac{P\pi\sigma\lambda^2}{P_{\min}64b^4}}, \]

so that, for a given beam width, it is preferable to operate at

long waves; but, of course, the dimensions of antennas having patterns of the same width increase as the wavelength is lengthened.

The minimum received power is determined by the level of the noise arising in the receiver, chiefly due to fluctuations; calculation shows that the inequality

\[ P_{\min} \ge n \cdot k \cdot T \cdot \Delta f \ \text{watts}, \]

must be satisfied, where \(k\) is Boltzmann’s constant, equal to \(1.38 \cdot 10^{-23}\ \dfrac{\text{watt}\cdot\text{sec}}{\text{degree}}\), \(T\) is the absolute temperature, \(n\) is a multiplier characterizing the noise, and \(\Delta f\) is the width of the frequency band passed by the receiver.

The optimum signal-to-noise ratio for the case of reception of electromagnetic pulses that is of interest to us is obtained when the condition

\[ \Delta f = \frac{1}{\tau_1}, \]

is observed, where \(\tau_1\) is the pulse duration. Taking this circumstance into account, we obtain

\[ R_{\max} \le \sqrt[4]{\frac{P \tau_1 \sigma A^2}{4 \pi n k T \lambda^2}} \tag{3} \]

But \(P\tau_1\) is equal to the energy of the pulse, which thus determines the maximum range. Since at large ranges \(\tau_1\) can be increased, long-range radars, in terms of power, do not differ very greatly from short-range radars.

It also follows from the last relation that a simple increase in power does not give a significant increase in range; thus, doubling the power increases the range by only 19%.

From equation (3) an interesting consequence is readily obtained. The relation between the maximum range of a radar and the equivalent reflecting surface \(\sigma\) for two objects of different sizes is, evidently, as follows:

\[ \frac{R_1}{R_2} = \sqrt[4]{\frac{\sigma_1}{\sigma_2}}, \]

where the subscripts 1 and 2 refer respectively to the first and second objects.

If for an airplane we take \(R_2 = 300\ \text{km}\) and \(\sigma_2 = 5\ \text{m}^2\), then an object located at a distance \(R_1 = 400 \cdot 10^3\ \text{km}\) can be detected if its equivalent reflecting surface is equal to:

\[ \sigma_1 = \sigma_2 \left(\frac{R_1}{R_2}\right)^4 = 15.8 \cdot 10^{12}\ \text{m}^2. \]

Taking the reflecting surface to be a circle, we obtain for its radius \(\rho\) the value

\[ \rho = \sqrt{\frac{\sigma_1}{\pi}} = 2.24 \cdot 10^6\ \text{m} = 2240\ \text{km}, \]

which agrees rather closely with the radius of the Moon (1739 km), located at a somewhat smaller distance from the Earth (384,400 km). Thus, one may hope to measure the distance to the Moon by using existing radars. Their range of action can be increased by lengthening the duration of the pulses; since in this case the delay time of the reflected signal will be of the order of 2 seconds, the pulse frequency must likewise be considerably reduced (see equation 1), and therefore overloading of the generator will not occur.

Indeed, in the winter of 1945–1946 the Americans reported that they had succeeded in “detecting” the Moon by means of radar and in measuring the distance to it with an accuracy not inferior to astronomical measurements. If one makes an analogous approximate calculation for a body situated at a distance of \(150 \cdot 10^6\ \text{km}\), it turns out that one may hope to detect a body whose radius exceeds the radius of the Sun by approximately 50 times; therefore radar location of the Sun can hardly be carried out at the present time.

Pulse generators have found broad application in investigations of the ionosphere, which are essentially also radar investigations. However, ionospheric radio engineering proved insufficient for solving the problems of military radar, which required a considerable improvement of the existing ionospheric installations with respect to increasing pulse power, improving antenna directivity, and substantially increasing the frequency of oscillation and the sensitivity of receiving devices. An excellent illustration of the exceptional sensitivity of modern receiving devices is the recent report\(^6\) by one of the most prominent American investigators in the field of microwaves, Southworth, that he succeeded in detecting solar radiation in the centimeter-wave region (between 1 and 10 cm), with one of the three measured values of the radiation energy agreeing very well with the value calculated from Planck’s radiation formula (if the Sun is likened to a black body with a temperature of \(6000^\circ\)), while the other two differed from the theoretical value by 12 and 80% respectively.

Details of Southworth’s measurements are not reported. But if one takes into account that the radiation energy in the range in which these measurements were made is almost a million times smaller than in the frequency region corresponding to the maximum radiation of the Sun, then one may estimate the exceptional sensitivity of the apparatus he used, which apparently is capable of registering a power of the order of \(10^{-15}\) watt.

The task of determining the distance to an aircraft is the simplest. Its solution is carried out by the scheme of Fig. 2.

The transmitter emits pulses whose repetition frequency and duration are regulated by a time-base generator, which simultaneously supplies to the plates of the capacitor of the electronic oscilloscope a voltage varying linearly with time and displacing the electron beam horizontally. The pulse reflected from the aircraft, after passing through the receiving device, produces a voltage pulse on the plates of the second capacitor of the oscilloscope, deflecting the electron beam vertically. As a result, two pulses appear on the screen (Fig. 3)—from the directly received and the reflected signals. The distance between them determines the delay time of the reflected pulse and, consequently, the distance to the aircraft. The oscilloscope scale can be graduated directly in kilometers of distance to the aircraft.

Fig. 2.

For a more accurate determination of distance, this simple scheme requires improvements that give a very significant effect.

One of the simplest and highly effective ways of increasing the accuracy of range measurement is the following: to the plates of the capacitor creating the horizontal sweep, in addition to the voltage proportional to time, a constant voltage is applied, allowing the entire sweep line to be shifted to the left. The magnitude of this voltage can be varied continuously. The reflected signal obtained on the oscilloscope screen is shifted, using the indicated constant voltage, until it coincides with a sighting thread stretched in front of the screen. The magnitude of the shifting voltage depends on the delay of the reflected signal, and the positions of the control knob can be graduated directly in kilometers of distance to the target.

Fig. 3.

Determination of azimuth is a more complicated task. Since the receiving antenna has a directional action characterized by the polar diagram of Fig. 4, a, in the simplest case the operator rota-

brings it to the point where the signal reflected by the aircraft and observed on the oscilloscope screen becomes most intense. The aircraft is then in the direction \(OB\). But, since at not too short wavelengths the width of the pattern, i.e. the magnitude of the angle between the maximum and half-value of the radiation field (or energy), is rather large (\(\beta\) in Fig. 4, \(a\)), the accuracy of such a determination is insufficient. It can be increased in various ways. When there are two identical receiving antennas whose axes form a certain angle (Fig. 4, \(b\)), an automatic switch is used which alternately feeds to the oscilloscope the voltage from one antenna and from the other, and slightly shifts (during one of the switchings) the phase of the horizontal sweep voltage. Since the pulses follow one another very frequently (in the radar mentioned above the number of pulses per second is 4100), the operator sees on the oscilloscope screen at once both deflections of the beam, located almost side by side, caused by the two antennas.

In the case of Fig. 4, \(b\), the aircraft is in the direction \(OB\); the right-hand antenna receives a stronger pulse (proportional to the segment \(OF\)) than the left-hand one (the segment \(OG\)). When the antennas are rotated about the vertical axis to the position corresponding to Fig. 4, \(c\), the stronger pulse will be received by the left-hand antenna. If, however, the antennas are set so that both pulses become equal, then, obviously, the aircraft will be in the direction \(OB\) (Fig. 4, \(d\)). The accuracy of this setting is considerably higher than in the case of a single antenna.

Fig. 4.

At comparatively low frequencies the antennas become cumbersome, and it is difficult to rotate them. In this case the determination of the azimuth is possible according to the principle of the radio direction finder.

A radio direction finder is, in principle, a flat coil rotated about an axis. The intensity of reception by such a coil has a maximum if its plane coincides with the direction of propagation of the radio wave (since in this case, in the sides of the coil parallel

to the vector of the electric field, opposing emfs are induced, shifted in phase by an angle corresponding to the distance between these sides). If, however, the plane of the loop is placed parallel to the wave front, then the phase difference disappears and the sum of the induced emfs becomes zero. Usually the orientation is carried out according to the minimum intensity of reception. To increase the reliability of the measurement and to increase the sensitivity of reception, antennas arranged in the horizontal plane in mutually perpendicular directions are used; the signals received by them are amplified and fed to a radiogoniometer.

In an analogous way the angle of elevation can be determined; however, its determination at large distances presents considerable difficulties, since at large distances the value of the angle of elevation is, obviously, small, and the wave reflected by the aircraft may reach the receiver both directly and by reflection from the surface of the Earth, which may create errors if this circumstance is not taken into account.

In the newest radars the method described has been considerably improved and gives excellent results. Thus, for example, in some American radars, in order to simplify the construction, the antenna determining the direction is divided into two parts \(A\), \(B\) (Fig. 5,a), connected by a phase regulator \(F\) (a line whose length is less than half the length of the working wave). The signal is received alternately from each half of the antenna, which is achieved by using a switch operating with a number of switchings equal to the number of pulses emitted per second. When the signal is taken from part \(A\) of the antenna, the signal received by part \(B\) is delayed by a certain interval of time \(T\), which is equivalent to the delay obtained in the case where the phase regulator is absent (Fig. 5,b), but the antenna forms with the front of the incoming wave some spatial angle \(\alpha\), owing to which part \(B\) is excited later than \(A\) by the interval of time required for the wave to traverse the segment \(l\).

Fig. 5.

When switching, the roles of the antennas are interchanged, which is equivalent to rotating the antenna (without a phase regulator) in the opposite direction through the same angle.

Owing to the high switching frequency, both received signals are visible at once on the oscilloscope screen, their intensities being equalized by the operator by rotating the antenna.

By changing the length of the phase regulator, it is possible to change the angle \(2\alpha\) between the axes of the antennas, which also offers known advantages.

2. LONG-RANGE RADARS

Long-range radars (detection stations) are intended for detecting the enemy at great distances, which makes it possible to bring air defense into action in good time. Since they are not required to correct fire, the accuracy of range and azimuth determination need not be very high. The angle of elevation may not be determined at all (at least at great ranges).

In the USA, long-range radars began to be designed after the development of short-range radars (gun-laying stations), approximately from 1938. Therefore the setting up of production was carried out in record time; from the manufacture of a laboratory model to the start of series production, undertaken by the Westinghouse firm with the participation of a number of the largest radio and electrical-engineering firms, only 6 months elapsed. By the time of the Japanese attack on Pearl Harbor, 112 radars had been produced, and their total production reached 788 units.

Fig. 6. General view of the mobile radar SCR-270/271.

The SCR-270/271 long-range radar consists of a pulsed transmitter with a power of 100–300 kW, operating on a wavelength of 2.70 m; the transmitter emits 621 pulses per second; the pulse duration is from 10 to 30 microseconds. Transmission and reception are carried out on a common antenna, whose radiation pattern has a width of 28° (in the horizontal plane) and 10° (in the vertical). The antenna rotates about a vertical axis, making 6 revolutions per minute and making it possible to carry out a circular survey of the terrain. The radar range is nearly 200 km for bombers and about 130 km for fighters.

The accuracy of range determination is 5–7 km; azimuth is determined with an accuracy of up to 4°.

A general view of the mobile model of this radar is shown in Fig. 6, in which the rotating antenna, mounted on ...

... on a cart. On the left is a truck with a power unit and a second truck, where the operator with the oscilloscope is located.

The stationary radar SCR-271-D, differing from the preceding one by a more advanced observation system (the so-called circular-scan sweep, see below), is shown in Fig. 7. The antenna, mounted on a 35-meter tower, also rotates about a vertical axis. At the top of the tower the so-called identification antenna is visible, making it possible to distinguish one’s own aircraft from enemy aircraft. A small transmitter is installed on friendly aircraft; it is automatically actuated by the radar signal striking the aircraft and, in response, emits coded signals received at the radar installation; these signals, called normal signals, allow ground operators to distinguish their own aircraft from enemy ones. If an aircraft is damaged, the same transmitter sends another type of signal—a distress signal, drawing the attention of ground stations to the aircraft in distress.

Unlike the Americans, the British began work on radars with long-range installations. They used stationary, rather bulky installations that received the name CH stations (chain home—fence), installed along the coast. One of such stations is shown in Fig. 8. The azimuth was determined by means of a radio direction finder; to determine the flight altitude, interference of the waves reflected by the aircraft and reaching the receiving antenna by two paths was used: one wave comes directly from the aircraft, while the other undergoes an additional reflection from the sea surface. By comparing the interference patterns received by two antennas located at different heights, it is possible to measure the altitude of an aircraft located at a distance of 100 km with an accuracy of up to 500 meters. At greater distances the measurement becomes unreliable; the full range of action of a CH station reaches 170 km, and the accuracy of azimuth determination is 2°.

Fig. 7. Stationary radar SCR-271-D.

However, these stations did not make it possible to detect aircraft flying very low. The point is that the intensity of reception of a signal reflected from the Earth is proportional to the square of the product of the hei-

… the antenna height to the aircraft’s flight altitude, and inversely proportional to the square of the wavelength; moreover, the wavelength of the CH stations was fairly large.

In practice, it proved possible to use CH stations to detect aircraft flying at altitudes from 150 to 5000 m.

Special devices that made it possible to separate individual reflected signals from the general mass of them made it possible to determine—

Fig. 8. Stationary CH radar station.

—the total number of aircraft even in massed raids, in which 100–150 machines took part, with an accuracy of up to 10% of their number.

During the period of the Munich Agreement (September 1938), the radar defense of the English coasts was in an embryonic state—only the London area was equipped with radars, the range of which is shown by straight hatching in Fig. 9,a. By September 1939 the CH radar network covered the entire eastern coast of England; at the height of the Battle of Britain (September 1940) it had been extended to the northwest and southwest and supplemented by a network of CHL radars, operating at a higher frequency and making it possible to catch low-flying aircraft. By September 1941 the British Isles were protected on all sides by a double “barrier,” shown in Fig. 9,b, where the “barrier” of CH radars is shown by straight hatching (the outer barrier), and that of the CHL stations by cross-hatching (the inner belt of defense).

Fig. 9. Area of operation of English radar stations.

3. PLAN-POSITION INDICATOR SCAN

The system of horizontal scanning considered above, very convenient for observing the motion of an individual airplane flying in the direction of the radar, and also the motion of several airplanes whose azimuths are close in magnitude, proves to be of little use when searching for airplanes at different azimuths, since, with sufficiently rapid rotation of the radio beam, the pulses observed at a given azimuth are visible on the oscilloscope screen for a very short time, while pulses from other azimuths at that moment are not registered at all.

A more advanced scan is the plan-position indicator scan (in English terminology PPI—plan-position indicator), the essence of which is as follows: in the absence of a signal the electron beam falls at the center of the oscilloscope screen; at the moment the pulse is sent, it begins to move in the radial direction at constant speed and moves in this way throughout the entire duration of the pause between pulses, and then returns very rapidly back to the center. As the antenna rotates, whose beam “illuminates” different azimuths, the radial scan of the oscilloscope also rotates in the plane of the screen with the same angular velocity. Thus, in its motion the electron beam fills the area of a circle with its center in the middle of the oscilloscope screen. The signal reflected from an airplane is fed to the modulator of the electron beam and changes its speed or intensity, as is done in electronic television receivers. Owing to this, brighter (or darker) spots are obtained in certain parts of the field of view, indicating the presence of an airplane (or another target). The afterglow time of the oscilloscope screen is chosen so that the operator sees the entire picture simultaneously. As a result, the screen shows a planar projection of the space surrounding the radar, with images of all targets detected by the radar. The target images are displaced from the center by distances proportional to the ranges of the targets; the azimuths of the targets are determined by means of the circular scale surrounding the luminous circle on the oscilloscope screen. The difference in brightness between the background and the targets can be made very considerable, owing to which the target images are clearly visible against the background of the screen. Nearby there is usually an indicator with controlled horizontal scanning; on its screen one can select one or several targets whose motion, for one reason or another, requires especially careful observation.

At meter wavelengths, whose antennas give a rather blunt radiation pattern, the outlines of objects are obtained to a considerable degree blurred; but when working with centimeter waves the antenna beam emitted can be made very sharp, and target images on the scan acquire great sharpness. In ca—

As an example, we present Fig. 10, which is a photograph of the oscilloscope screen of a centimeter-wave radar installed on an aircraft. The photograph was taken in the southwestern part of Wales near the towns of Pembroke and Milford Haven and is reproduced without any changes, except for the addition of place-name captions and a thin line characterizing the position of the true boundaries of the coastline of the area over which the aircraft was located at the moment of taking the picture.

Fig. 10. Image obtained on the screen of a centimeter-wave aircraft radar with a circular-scan sweep.

The appearance of the circular-scan sweep and of centimeter-band radars opened up extraordinary prospects for the use of radars under military and peacetime conditions (see below).

4. SHORT-RANGE RADARS

(gun-laying stations)

In contrast to long-range radars, which must detect a distant aircraft in good time and determine its location with comparatively crude accuracy, short-range radars must determine the coordinates of an aircraft with a high degree of accuracy. On the other hand, since their radius of action is much smaller than that of the preceding ones, the power of the generator may also be smaller; therefore it is possible to operate at higher frequencies, which makes it possible to reduce the dimensions of the antennas. The British, as indicated above, followed this path, creating a network of CHL radar stations, which were used in

...at short ranges. The antenna of the CHL station is visible in Fig. 11. Fig. 12 shows an antenna for measuring the elevation angle of an aircraft, used in British close-range radars operating in the centimeter band. Fig. 13 depicts an antenna serving a receiver with circular-scan sweep, and Fig. 14 shows the antenna of an aircraft-guidance station, used to determine the position of an enemy aircraft and to transmit its coordinates to a night fighter, which is thereby directed to the target and approaches it to within 3–5 km; after that it tracks the target by means of its own radar, installed on the fighter and having a short range. It is known that these latter installations are called by the British, respectively, GCI (Ground Control Interception, i.e., ground-controlled interception) and AI

Fig. 11. Antenna of the CHL station.

(Airborne Interception), which appeared in 1941, quickly gave the British superiority in the air. Thus, during one raid by German aviation on London, out of 500 bombers 185 were shot down.

British radars developed in the direction from long-range radars to short-range ones. By contrast, the Americans began working

Fig. 12. Elevation-angle antenna of a short-range radar operating on centimeter waves.

on short-range radars, and only later were the long-range radars described above developed. As a result of the work of research institutions of the United States armed forces, by November 1938 the SCR-268 radar had been created; it operated at an oscillation frequency of 110 to 240 megacycles (different models) and made it possible to detect aircraft at distances up to 40 km, determine the azimuth with an error up to 4°, the elevation angle with an accuracy up to 2°–5°, and the distance to the aircraft with an accuracy to—

N. N. MALOV

accuracy of up to 0.2 km. After certain improvements, this model was put into serial production. Fig. 15 shows the transmitter antenna, rotating about a vertical axis; in

Fig. 13. Receiver antenna with circular-scan sweep.

Fig. 16—the system for measuring the elevation angle, rotating about horizontal and vertical axes; in Fig. 17—the azimuthal

Fig. 14. Antenna of an aircraft-guidance station.

antenna, capable of rotating in the horizontal plane. Both of the latter antennas are dual antennas, and the axes of their patterns form a certain angle; for the precise determination of the coordinates of the aircraft, the pri-

Fig. 15. Antenna of the SCR-268 radar transmitter.

Fig. 16. Antenna for measuring the angle of elevation.

Fig. 17. Azimuth antenna.

Fig. 18. Oscilloscopes for determining three coordinates (SCR-268 radar).

Fig. 19. Mobile SCR-268 radar: 1 — azimuth antenna, 2 — range oscilloscope, 3 — filament transformer, 4 — transmitter, 5 — elevation-angle oscilloscope, 6 — radiating antenna, 7 — azimuth receiver, 8 — azimuth oscilloscope, 9 — concentric box, 10 — elevation-angle receiver, 11 — manipulator, 12 — modulator, 13 — elevation-angle antenna.

the “two-lobe method” described above is used (Figs. 4, 5). The radar has three oscilloscopes for determining three coordinates. The oscilloscopes are clearly visible in Fig. 18. The total number of electron tubes in such a radar is 110. A later model of the same radar is mounted on a common movable base (Fig. 19). The same installation in operation is shown in Fig. 20. It is serviced by five operators.

The coordinates measured by the radar are transmitted to a neighboring searchlight, which at the required moment is switched on and immediately illuminates the enemy aircraft, and also to an antiaircraft battery, where the proper firing data are at once set.

During tests in 1938, such a radar was not only capable of continuously tracking an aircraft, but also made it possible to regis-

Fig. 20. SCR-268 radar in operation.

ter the bursts of 3-inch antiaircraft shells at a distance of several kilometers. A skeleton diagram of the SCR-268 radar, in which solid lines indicate the electrical connections of the individual units of the installation, and dashed lines indicate mechanical linkages, is shown in Fig. 21. The hand wheels by means of which the operators rotate the corresponding antennas are mechanically connected with the automatic devices controlling the searchlights and antiaircraft guns. Thus, having acquired an aircraft, the radio operators keep it in the radar’s field of view, while the searchlights and guns continuously follow the motion of the aircraft and are brought into action at the required moment.

SCR-268 radars were tested in 1941 during maneuvers south of the Panama Canal; in August 1941 two radars were installed in Iceland to protect the northern route connecting America with England and the USSR; in December 1941, during the Japanese raid on

Pearl Harbor; 16 radars operated there. Subsequently they were used on all fronts. The total number of SCR-268 radars manufactured

Fig. 21. Block diagram of the SCR-268 radar.

reached the impressive figure of 2,974 units. Of course, their design could not remain secret, and in the middle of 1944 the Japanese completely copied this radar; but it later became clear that already in March 1944 it had been taken out of production by the Americans and replaced by a more advanced model.

Radar installations of an analogous type were installed on naval vessels and ensured the detection of targets in complete darkness or in fog, as well as reliable correction of fire, since they registered the fountains of water produced when artillery shells struck the water.

American radar installations of this kind (SCR-584 and others), operating at wavelengths of about 10 cm, made it possible to determine range with an accuracy of up to 10–12 m; the accuracy of determining direction reached 0°,06; the maximum radius of action was 30–40 km (depending on the size of the target). The number of radars of this type produced during the war amounted to 1710 units. These installations sharply increased the combat power of the Allied navies. Radars ensured the defeat of the Italian fleet in the Mediterranean, the destruction of the German battleship Scharnhorst in the darkness of the polar night in the Norwegian Sea, and the success of a number of major operations in the Pacific theater of war.

Shipborne and aircraft radars proved to be a powerful means of combating submarines.

In conclusion, mention should be made of a very ingenious method of reducing the effectiveness of enemy radars, used by the British. When aircraft approached the designated target, the British dropped strips of metallized paper from them; the reflections from these strips were so numerous that the operators of German radars were completely at a loss—it seemed to them that the entire sky was covered with Allied aircraft.

5. RADARS AS A MEANS OF NAVIGATION

It is quite obvious that a radar installed on the ground can serve as an excellent navigational aid, since it makes it possible to determine the coordinates of an aircraft or vessel and to give it instructions about the direction of further flight (or sailing).

The war showed, however, that much more could be done. During the first mass raids by Allied aviation on Germany, difficulties arose connected with the need to concentrate the greatest possible number of aircraft over a small area. It was necessary to ensure the possibility of error-free and rapid target finding, timely bombing, and the urgent departure of the formations that had completed their bombing.

The Allies developed a system for guiding aircraft to the target, which received (apparently because of its exceptional effectiveness) the peculiar name “Gee” (damn it!) and consisted of the following. Three transmitting stations, \(A\), \(B\), \(C\), are located on the ground, simultaneously sending out pulse signals (Fig. 22). The navigator of an aircraft, receiving two of these signals (for example, \(A\) and \(B\)) on an oscilloscope, cannot, of course, determine the distance of the aircraft to the stations, but he can easily determine how much farther one station is from him,

than another. Since the curves corresponding to a given difference in the distances to stations \(A\) and \(B\) are hyperbolas (for simplicity we shall consider the plane problem), whose foci coincide with the stations, then, using a previously prepared chart, the navigator determines on which of the hyperbolas the aircraft is located (let this be the hyperbola \(ab\)); however, he still cannot determine at which particular point of the hyperbola the aircraft is located. But by making an analogous observation of the signals from stations \(A\) and \(C\) (or \(B\) and \(C\)), the navigator determines the hyperbola \(ac\), passing through the position of the aircraft. Obviously, the intersection of the hyperbolas found uniquely determines the position of the aircraft (point \(S\)).

Fig. 22. The “Gee” system for guiding aircraft to a target.

The accuracy of this method proved to be exceptionally high.

At present, Gee stations and similar American “Loran” stations cover a considerable part of the world map with a navigation network, which will greatly facilitate air and sea navigation and will make it independent of weather conditions, since the navigator will no longer need to “catch” the sun or the stars in order to determine the position of a ship.

The American navigation stations “Loran” operate at frequencies from 1.7 to 2.0 megacycles and are capable of providing navigation at distances up to 2500 km.

In these stations the radio transmitter \(A\) (Fig. 22) is duplicated: one of the transmitters (\(A_1\)) gives 25 pulses per second. These pulses actuate the automatic station \(B\), which in turn begins to send pulses, evidently delayed with respect to the first by a certain constant interval of time. The ship’s navigator, receiving both pulses on an oscilloscope whose horizontal sweep frequency is equal to 25 hertz, determines his position on one of the hyperbolas \(ab\). The other transmitter \(A_2\) of station \(A\) excites station \(C\) in an analogous manner, but with a somewhat different pulse frequency (25.062 pulses per second). Owing to the difference in pulse frequency, the latter are not visible on the oscilloscope screen (since they are shifting all the time); but if the navigator changes the sweep frequency and makes it equal to 25.067 hertz, then the pulses from stations \(A_2\) and \(C\) will be visible on the screen, whereby the navigator will determine the hyperbola \(ac\) on which the ship is located. The two measurements make it possible to determine the true position of the ship (the point of intersection of the hyperbolas).

The accuracy of location given by Gee installations at a distance of 500 km from the stations is on the average several square kilometers,

increasing, for cases when the aircraft is near the middle of the straight line connecting the two stations, to 0.75 km.

However, even this surprisingly high accuracy proved insufficient for mass precision bombing of small targets, and a masterpiece of modern radar technology was created: the “Oboe” system. This system likewise has 2 ground stations and a pre-prepared navigation grid.

They simultaneously send out pulses, received by the aircraft radio installation and automatically triggering the transmitter of the aircraft’s recognition signal; this recognition pulse is received by the stations and transmitted to an oscilloscope, making it possible to determine its relative delay in comparison with the pulses of the ground stations and, consequently, the position of the aircraft. The aircraft receives instructions about its further course and is guided onto a hyperbola passing through the bombing target. As it approaches the target, the stations notify the aircraft that 5 minutes, 3 minutes, 1 minute remain to the target (since the station knows the coordinates of the aircraft and its speed).

At the required moment, a signal is given to release the bombs (which, incidentally, can also be done automatically by means of a remotely controlled bomb release). Throughout this entire operation, the personnel of the ground stations know the aircraft’s position more accurately than its crew does.

At a distance of 350 km from the stations, the error in determining the aircraft’s coordinates with this target-guidance system does not exceed 200 m (under combat conditions); in peacetime conditions it can be reduced severalfold more.

The system described was tested, in particular, during the Battle of the Ruhr basin.

An aircraft equipped with a radar installation with all-round scanning, operating on centimeter waves, is able to see on the oscilloscope screen objects as small in size as submarines and individual ground structures. Thanks to this, exceptionally reliable precision bombing is possible.

There is a known case in which, when it was necessary to enable persons imprisoned in a jail to escape confinement, aircraft bombed the guardroom and the walls, leaving intact the part of the building in which the prisoners were located.

Cases are also described of successful precision bombing of specified houses while neighboring buildings remained relatively intact. Very accurate determination of the coordinates of a flying object, in combination with fighter target guidance, played a major role in the successful struggle of British aviation against flying bombs.

Finally, mention should be made of radar shells. At the end of the war the British produced antiaircraft artillery shells equipped with

miniature radar installation; the pulses of this radar, reflected from the aircraft under fire and received by the projectile’s radar, actuated a detonator, which ensured the bursting of the projectile at a specified distance from the aircraft, calculated in such a way that the probability of the aircraft being struck by fragments of the projectile was very great.

In conclusion it is not without interest to cite certain data characterizing the various types of radar installations produced by American industry during the war years. These data excellently

illustrate the variety of types and the exceptional qualities of these remarkable installations. Of the 10 types that have been described in print, we give information on only 5 types, those most different from one another.

6. RADIO SERVICING OF MODERN COMBAT AIRCRAFT

The manifold applications of radio-engineering installations on modern combat aircraft lead to the complication of the radio equipment of aircraft and require the creation of an entire network of ground radio stations servicing the aircraft in flight. The complexity of the radio equipment of a heavy aircraft may be illustrated by the following data: the number of radio-engineering units (transmitters, receivers, amplifiers, etc.) on an aircraft of the “Lancaster” type reaches 50,

the number of tubes in them is more than three hundred, the number of different antennas is more than 10.

Figure 23 schematically shows the radio-service system of the “Lancaster” aircraft.

1 — Beneath the recognition marking depicted on the fuselage of the aircraft is located a radar with circular-sweep scanning for detailed survey of the terrain over which the flight is being made.

Fig. 23. Radio-service system of the “Lancaster” aircraft.

2 — The aircraft maintains telegraph communication with command.

3 — “Gee” equipment, making it possible to direct the aircraft’s movement from the ground.

4 — Radio beacons serving the aircraft on its return.

5 — A radio beacon operating on medium waves facilitates a rough determination by the aircraft of the direction of flight; reception of its signals is carried out by a loop system.

6 — A radio beacon regulating the aircraft’s movement when approaching the aerodrome. It consists of two directional radio transmitters, one of which transmits telegraph signals corresponding to the letter A, the other to the letter N, with the transmission of one transmitter occurring during the pauses of the other (Fig. 24). If the aircraft is flying farther to the right or to the left than it should, the pilot hears the transmission of one or the other letter more loudly; but if

Fig. 24.

if the course is correct, then the aircraft enters the “equisignal” zone, where both letters are received with equal intensity, so that the pilot hears a continuous sound of constant strength (this method of indicating the path is essentially the same as the two-petal method described above, Fig. 4).

7—Signaling that regulates the landing of the aircraft by means of antenna 11; the ground microwave transmitter is connected by telephone communication with the auxiliary high-frequency station 8. If this communication is disrupted because of damage to the aircraft, then an auxiliary system comes into operation, working at a lower frequency and received by the aircraft on an antenna located above the fuselage.

9 and 10—Radio beacons, one of which is installed, approximately, 3 km from the aerodrome, the other at the boundary of the aerodrome. The signals sent by them allow the pilot to determine the boundaries of the aerodrome and to begin landing on the correct glide path, coinciding with the equisignal zone of the radio beacons.

12—Communication between aircraft in flight.

13—A medium-wave-band station with a large range of action (1000–1200 km), received by the aircraft on a trailing antenna, unwound at the necessary moments from a special drum.

14—The system of aircraft recognition signals—normal and emergency (see § 2).

15—Protective radar, announcing the appearance of an enemy fighter in the tail of the aircraft.

16—A system of sound signaling, set in operation by ground radio stations installed in mountainous terrain if the aircraft approaches the mountains. The English figuratively call it the “mountain goat.”

This list clearly shows how diverse are the types of application of radio installations in modern air navigation, and makes it possible to assess the breadth of the necessary organizational work.

Such are the brief reports on the successes of radar technology which have now become the property of the general public. Undoubtedly, radars possess a whole series of valuable properties, opening broad prospects for their application not only under combat conditions, but also in peacetime.

LITERATURE

1. Colton—PIRE, 33, 740, 1945.
2. Watson—Wat. Discovery, 6, 281, 1945.
3. Electronics, 4, 9, 10 and 11, 1945.
4. Radio News, 6, 8, 1945.
5. Aviation, 6, 1915.
6. Southworth, Journ. Frankl. Inst., 6, 239, 285, 1945.